In the proposed research we will: (1) Document awareness, knowledge, and perceptions of NORPLANT among nationally representative samples of women and men in the U.S., and examine subgroup differences and correlates of variation among subgroups; (2) Model the formation of NORPLANT use intentions among women, and examine the relationship between attitudes and intention using the Theory of Reasoned Action; (3) Examine the variation in the women's intention to use NORPLANT, and men's attitudes toward NORPLANT use by their spouse/partner; (4) Estimate the incidence and prevalence of NORPLANT adoption; model the predictors/determinants of adoption using a social-demographic model, and examine subgroup differences; (5) Test behavioral models to investigate the decision making process leading to NORPLANT adoption, and examine the intention-behavior relationship; (6) Analyze NORPLANT adoption as a contraceptive switching process either from no method, or from another method, and to examine use of NORPLANT for the purpose of delaying vs terminating childbearing; (7) Examine change in knowledge, perceptions, intentions, and use over a period of four years; and (8) Estimate continuation and discontinuation rates using multivariate (with covariates) hazards models. The first three aims will be accomplished with extant data from the two national surveys we have conducted in 1991. While the remaining aims can also be accomplished with these data, their value will be greatly enhanced if we used longitudinal data to accomplish these. Hence we propose to conduct two follow-up interviews with the female baseline sample, one in 1992, and the other in 1993. All other data collection either have been, or will be, performed under existing grants. The research is guided by Fishbein and Ajzen's Theory of Reasoned Action, as well as by a Social-Demographic Model. The analysis includes descriptive analyses in the form of frequency distributions and multi-way cross-tabulations; multivariate analyses using logistic regression, and polytomous logit regression, and multivariate hazards modelling. At the end of each project year, the results will be disseminated through reports, working papers, presentations at professional meetings; and a final monograph will synthesize the results from all phases of the proposed research.